Overexpression of tumor necrosis factor alpha by a recombinant rabies virus attenuates replication in neurons and prevents lethal infection in mice

The effect of tumor necrosis factor alpha (TNF-alpha) on rabies virus (RV) infection of the mouse central nervous system (CNS) was studied, using recombinant RV engineered to express either soluble TNF-alpha [SPBN-TNF-alpha+] or insoluble membrane-bound TNF-alpha [SPBN-TNF-alpha(MEM)]. Growth curves derived from infections of mouse neuroblastoma NA cells revealed significantly less spread and production of SPBN-TNF-alpha+ than of SPBN-TNF-alpha(MEM) or SPBN-TNF-alpha-, which carries an inactivated TNF-alpha gene. The expression of soluble or membrane-bound TNF-alpha was not associated with increased cell death or induction of alpha/beta interferons. Brains of mice infected intranasally with SPBN-TNF-alpha+ showed significantly less virus spread than did mouse brains after SPBN-TNF-alpha- infection, and none of the SPBN-TNF-alpha+-infected mice succumbed to RV infection, whereas 80% of SPBN-TNF-alpha- -infected mice died. Reduced virus spread in SPBN-TNF-alpha+-infected mouse brains was paralleled by enhanced CNS inflammation, including T-cell infiltration and microglial activation. These data suggest that TNF-alpha exerts its protective activity in the brain directly through an as yet unknown antiviral mechanism and indirectly through the induction of inflammatory processes in the CNS.     

Biological basis of rabies virus neurovirulence in mice: comparative pathogenesis study using the immunoperoxidase technique.

The CVS strain of fixed rabies virus causes acute, fatal encephalomyelitis in young adult ICR mice. Variant RV194-2, which was selected from CVS virus in cell culture with a neutralizing antiglycoprotein monoclonal antibody, has a single amino acid change in the glycoprotein. The infections caused by CVS virus and RV194-2 virus were compared in mice for 14 days postinoculation of 5 x 10(7) PFU into the right masseter muscle. All CVS virus-infected mice died (mean time to death, 7.9 days), compared with a mortality rate of 8.5% for RV194-2 virus-infected mice. RV194-2 virus spread to the ipsilateral trigeminal ganglion during the first 2 days postinoculation, and both viruses spread to the ipsilateral motor nucleus of the trigeminal nerve in the pons. Both viruses spread centrifugally and caused infection of bilateral trigeminal ganglia on day 3. The viruses spread throughout the central nervous system (CNS) at similar rates, but CVS virus infected many more neurons than did RV194-2 virus. Rabies virus antigen was observed in only occasional CNS neurons after day 6 of RV194-2 virus infection. By this time, CVS virus had caused severe widespread infection. In this model, virulence depends on improved efficiency of viral spread between CNS neurons rather than the rate of spread or topographical distribution of the infection.     

The glycoprotein and the matrix protein of rabies virus affect pathogenicity by regulating viral replication and facilitating cell-to-cell spread

While the glycoprotein (G) of rabies virus (RV) is known to play a predominant role in the pathogenesis of rabies, the function of the RV matrix protein (M) in RV pathogenicity is not completely clear. To further investigate the roles of these proteins in viral pathogenicity, we constructed chimeric recombinant viruses by exchanging the G and M genes of the attenuated SN strain with those of the highly pathogenic SB strain. Infection of mice with these chimeric viruses revealed a significant increase in the pathogenicity of the SN strain bearing the RV G from the pathogenic SB strain. Moreover, the pathogenicity was further increased when both G and M from SB were introduced into SN. Interestingly, the replacement of the G or M gene or both in SN by the corresponding genes of SB was associated with a significant decrease in the rate of viral replication and viral RNA synthesis. In addition, a chimeric SN virus bearing both the M and G genes from SB exhibited more efficient cell-to-cell spread than a chimeric SN virus in which only the G gene was replaced. Together, these data indicate that both G and M play an important role in RV pathogenesis by regulating virus replication and facilitating cell-to-cell spread.     

Differentiation of an adult neuron cell line increases susceptibility to rabies infection

A wide variety of in vitro models have been used for studying rabies infection, however, currently, no central nervous system (CNS) adult neuron cultures are available. The current study determined the susceptibility to rabies infection in an adult CNS neuron cell line (CAD-R1). Cultures of CAD-R1 cells were held for 5 days in medium containing serum (undifferentiated CAD-R1 cells) or in serum-free medium (differentiated CAD-R1 cells). They were then infected with highly neurotropic rabies virus (RV) strain (CVS), obtained from fibroblastic cells (CVS-BHK) or from adult mouse brain (CVS-MB). Undifferentiated and differentiated cells were infected with the two RV strains, but the percentage of infected cells in differentiated cultures was significantly greater (83% and 79%, respectively) than in undifferentiated cells (51% and 60%) (Student's t test<0.05). Susceptibility to infection apparently depended on cellular differentiation state, possibly due to acquisition of additional morphological and biochemical characteristics during the differentiation process that made them more susceptible to RV infection. Therefore, CAD R1 cells may represent a good model for RV infection, making them a useful tool for studying RV neurotropism, infection pathogeny, isolation of street virus or producing safer and most potent vaccines.     

Apoptosis and rabies virus neuroinvasion

Rabies virus (RV) causes a non-lytic infection of neurons leading to a fatal myeloencephalitis in mammals including humans. By comparing the infection of the nervous system of mice by a highly pathogenic neuroinvasive strain of RV (CVS) and by a strain of attenuated pathogenicity (PV) with restricted brain invasion, we showed that RV neuroinvasiveness results of three factors: not only neurotropic RV avoids induced neuron cell death but also "protective" T cells that migrate into the infected nervous system are killed by apoptosis and finally inflammation of the infected nervous system is limited. Our data suggest that the preservation of the neuronal network, the limitation of the inflammation and the destruction of T cells that invade the CNS in response to the infection are crucial events for RV neuroinvasion and for transmission of RV to another animal.     

Pathways of the early propagation of virulent and avirulent rabies strains from the eye to the brain.

Penetration of the central nervous system of the adult rat by the CVS strain of rabies virus and its two avirulent derivatives Av01 and Av02 has been studied by inoculation of the virus into the anterior chamber of the eye. The primary sites of penetration of CVS were (i) the intraocular parasympathetic oculomotor fibers, (ii) the retinopetal fibers of pretectal origin, and (iii) the intraocular fibers of the ophthalmic nerve. The mutant strains, however, lost the capacity to invade the two former groups of fibers, although their penetration into the trigeminal system was not impaired. Neither strain CVS nor the mutants infected primarily the intraocular adrenergic terminals and the optic nerve. Mutant strains, but not CVS, were able to infect the lens. These results indicate that the cholinergic receptor may not be the only receptor for rabies virus and that rabies virus is conveyed in the nervous system by retrograde axoplasmic flow. Strain CVS spread throughout the brain and propagated eventually back to the retina. The mutants penetrated the brain as well, but the infection was slow, involved different cerebral structures, and cleared up completely in 3 weeks, probably because of an efficient immune response.     

Rabies virus infection of cultured rat sensory neurons.

The axonal transport of rabies virus (challenge virus strain of fixed virus) was studied in differentiated rat embryonic dorsal root ganglion cells. In addition, we observed the attachment of rabies virus to neuronal extensions and virus production by infected neurons. A compartmentalized cell culture system was used, allowing infection and manipulation of neuronal extensions without exposing the neural soma to the virus. The cultures consisted of 60% large neuronal cells whose extensions exhibited neurofilament structures. Rabies virus demonstrated high binding affinity to unmyelinated neurites, as suggested by assays of virus adsorption and immunofluorescence studies. The rate of axoplasmic transport of virus was 12 to 24 mm/day, including the time required for internalization of the virus into neurites. The virus transport could be blocked by cytochalasin B, vinblastine, and colchicine, none of which negatively affected the production of virus in cells once the infection was established. It was concluded that, for the retrograde transfer of rabies virus by neurites from the periphery to the neuronal soma, the integrity of tubulin- and actin-containing structures is essential. The rat sensory neurons were characterized as permissive, moderately susceptible, but low producers of rabies virus. These neurons were capable of harboring rabies virus for long periods of time and able to release virus into the culture medium without showing any morphological alterations. The involvement of sensory neurons in rabies virus pathogenesis, both in viral transport and as a site for persistent viral infection, is discussed.     

The reovirus sigma1s protein is a determinant of hematogenous but not neural virus dissemination in mice

Nonstructural protein σ1s is a critical determinant of hematogenous dissemination by type 1 reoviruses, which reach the central nervous system (CNS) by a strictly blood-borne route. However, it is not known whether σ1s contributes to neuropathogenesis of type 3 reoviruses, which disseminate by both vascular and neural pathways. Using isogenic type 3 viruses that vary only in σ1s expression, we observed that mice survived at a higher frequency following hind-limb inoculation with σ1s-null virus than when inoculated with wild-type virus. This finding suggests that σ1s is essential for reovirus virulence when inoculated at a site that requires systemic spread to cause disease. Wild-type and σ1s-null viruses produced comparable titers in the spinal cord, suggesting that σ1s is dispensable for invasion of the CNS. Although the two viruses ultimately achieved similar peak titers in the brain, loads of wild-type virus were substantially greater than those of the σ1s-null mutant at early times after inoculation. In contrast, wild-type virus produced substantially higher titers than the σ1s-null virus in peripheral organs to which reovirus spreads via the blood, including the heart, intestine, liver, and spleen. Concordantly, viral titers in the blood were higher following infection with wild-type virus than following infection with the σ1s-null mutant. These results suggest that differences in viral brain titers at early time points postinfection are due to limited virus delivery to the brain by hematogenous pathways. Transection of the sciatic nerve prior to hind-limb inoculation diminished viral spread to the spinal cord. However, wild-type virus retained the capacity to disseminate to the brain following sciatic nerve transection, indicating that wild-type reovirus can spread to the brain by the blood. Together, these results indicate that σ1s is not required for reovirus spread by neural mechanisms. Instead, σ1s mediates hematogenous dissemination within the infected host, which is required for full reovirus neurovirulence.     

Herpes simplex virus type 1 glycoprotein e is required for axonal localization of capsid, tegument, and membrane glycoproteins

Herpes simplex virus type 1 (HSV-1) glycoprotein E (gE) promotes cell-to-cell spread at basolateral surfaces of epithelial cells, but its activity in neurons is less clear. We used the mouse retina infection model and neuronal cell cultures to define the spread phenotype of gE mutant viruses. Wild-type (WT) and gE-null (NS-gEnull) viruses both infected retina ganglion cell neurons; however, NS-gEnull viral antigens failed to reach the optic nerve, which indicates a defect in axonal localization. We evaluated two Fc receptor-negative gE mutant viruses containing four amino acid inserts in the gE ectodomain. One mutant virus failed to spread from the retina into the optic nerve, while the other spread normally. Therefore, the gE ectodomain is involved in axonal localization, and the Fc receptor and neuronal spread are mediated by overlapping but distinct gE domains. In the retina infection model, virus can travel to the brain via the optic nerve from presynaptic to postsynaptic neurons (anterograde direction) or via nerves that innervate the iris and ciliary body from postsynaptic to presynaptic neurons (retrograde direction). WT virus infected the brain by anterograde and retrograde routes, whereas NS-gEnull virus failed to travel by either pathway. The site of the defect in retrograde spread remains to be determined; however, infection of rat superior cervical ganglia neurons in vitro indicates that gE is required to target virion components to the axon initial segment. The requirement for gE in axonal targeting and retrograde spread highlights intriguing similarities and differences between HSV-1 and pseudorabies virus gE.     

Different Patterns of Neuronal Infection after Intracerebral Injection of Two Strains of Pseudorabies Virus

Pseudorabies virus (PRV), a swine neurotropic alphaherpesvirus, is known to invade the central nervous system (CNS) of a variety of animal species through peripherally projecting axons, replicate in the parent neurons, and then pass transsynaptically to infect other neurons of a circuit. Studies of the human pathogen herpes simplex virus type 1 have reported differences in the direction of transport of two strains of this virus after direct injection into the primate motor cortex. In the present study we examined the direction of transport of virulent and attenuated strains of PRV, utilizing injections into the rat prefrontal cortex to evaluate specific movement of virus through CNS circuitry. The data demonstrate strain-dependent patterns of infection consistent with bidirectional (anterograde and retrograde) transport of virulent virus and unidirectional (retrograde) transport of attenuated PRV from the site of injection. The distribution of infected neurons and the extent of transsynaptic passage also suggest that a release defect in the attenuated strain reduces the apparent rate of viral transport through neuronal circuitry. Finally, injection of different concentrations of virus influenced the onset of replication within a neural circuit. Taken together, these data suggest that viral envelope glycoproteins and virus concentration at the site of injection are important determinants of the rate and direction of viral transport through a multisynaptic circuit in the CNS.     

Failure to open the blood-brain barrier and deliver immune effectors to central nervous system tissues leads to the lethal outcome of silver-haired bat rabies virus infection

Rabies is a lethal disease caused by neurotropic viruses that are endemic in nature. When exposure to a potentially rabid animal is recognized, prompt administration of virus-neutralizing antibodies, together with active immunization, can prevent development of the disease. However, once the nonspecific clinical symptoms of rabies appear conventional postexposure treatment is unsuccessful. Over the last decade, rabies viruses associated with the silver-haired bat (SHBRV) have emerged as the leading cause of human deaths from rabies in the United States and Canada as a consequence of the fact that exposure to these viruses is often unnoticed. The need to treat SHBRV infection following the development of clinical rabies has lead us to investigate why the immune response to SHBRV fails to protect at a certain stage of infection. We have established that measurements of innate and adaptive immunity are indistinguishable between mice infected with the highly lethal SHBRV and mice infected with an attenuated laboratory rabies virus strain. While a fully functional immune response to SHBRV develops in the periphery of infected animals, the invasion of central nervous system (CNS) tissues by immune cells is reduced and, consequently, the virus is not cleared. Our data indicate that the specific deficit in the SHBRV-infected animal is an inability to enhance blood-brain barrier permeability in the cerebellum and deliver immune effectors to the CNS tissues. Conceivably, at the stage of infection where immune access to the infected CNS tissues is limited, either the provision or the development of antiviral immunity will be ineffective.     

Therapeutic immune clearance of rabies virus from the CNS

The long-held concept that rabies infection is lethal in humans once the causative rabies virus has reached the CNS has been called into question by the recent survival of a number of patients with clinical rabies. Studies in animal models provide insight into why survival from a rabies virus infection that has spread to the CNS is possible and the immune mechanisms involved. In the CNS, both innate mechanisms capable of inhibiting virus replication and the activity of infiltrating rabies virus-specific T and B cells with the capacity to clear the virus are required. Deficiencies in the induction of either aspect of rabies immunity can lead to lethal consequences but may be overcome by novel approaches to active and passive immunization.     

Entry versus blockade of brain infection following oral or intraperitoneal scrapie administration: role of prion protein expression in peripheral nerves and spleen

Naturally occurring transmissible spongiform encephalopathy (TSE) diseases such as bovine spongiform encephalopathy in cattle are probably transmitted by oral or other peripheral routes of infection. While prion protein (PrP) is required for susceptibility, the mechanism of spread of infection to the brain is not clear. Two prominent possibilities include hematogenous spread by leukocytes and neural spread by axonal transport. In the present experiments, following oral or intraperitoneal infection of transgenic mice with hamster scrapie strain 263K, hamster PrP expression in peripheral nerves was sufficient for successful infection of the brain, and cells of the spleen were not required either as a site of amplification or as transporters of infectivity. The role of tissue-specific PrP expression of foreign PrP in interference with scrapie infection was also studied in these transgenic mice. Peripheral expression of heterologous PrP completely protected the majority of mice from clinical disease after oral or intraperitoneal scrapie infection. Such extensive protection has not been seen in earlier studies on interference, and these results suggested that gene therapy with mutant PrP may be effective in preventing TSE diseases.     

The range and distribution of murine central nervous system cells infected with the gamma(1)34.5- mutant of herpes simplex virus 1.

Wild-type herpes simplex virus 1 (HSV-1) multiplies, spreads, and rapidly destroys cells of the murine central nervous system (CNS). In contrast, mutants lacking both copies of the gamma(1)34.5- gene have been shown to be virtually lacking in virulence even after direct inoculation of high-titered virus into the CNS of susceptible mice (J. Chou, E. R. Kern, R. J. Whitley, and B. Roizman, Science 250:1262-1266, 1990). To investigate the host range and distribution of infected cells in the CNS of mice, 4- to 5-week-old mice were inoculated stereotaxically into the caudate/putamen with 3 x 10(5) PFU of the gamma(1)34.5- virus R3616. Four-micrometer-thick sections of mouse brains removed on day 3, 5, or 7 after infection were reacted with a polyclonal antibody directed primarily to structural proteins of the virus and with antibodies specific for neurons, astrocytes, or oligodendrocytes. This report shows the following: (i) most of the tissue damage caused by R3616 was at the site of injection, (ii) the virus spread by retrograde transport from the site of infection to neuronal cell nuclei at distant sites and to ependymal cells by cerebrospinal fluid, (iii) the virus infected neurons, astrocytes, oligodendrocytes, and ependymal cells and hence did not discriminate among CNS cells, (iv) viral replication in some neurons could be deduced from the observation of infected astrocytes and oligodendrocytes at distant sites, and (v) infected cells were being efficiently cleared from the nervous system by day 7 after infection. We conclude that the gamma(1)34.5- attenuation phenotype is reflected in a gross reduction in the ability of the virus to replicate and spread from cell to cell and is not due to a restricted host range. The block in viral replication appears to be a late event in viral replication.     

Presence of Virus Neutralizing Antibodies in Cerebral Spinal Fluid Correlates with Non-Lethal Rabies in Dogs

Background

Rabies is traditionally considered a uniformly fatal disease after onset of clinical manifestations. However, increasing evidence indicates that non-lethal infection as well as recovery from flaccid paralysis and encephalitis occurs in laboratory animals as well as humans.

Methodology/Principal Findings

Non-lethal rabies infection in dogs experimentally infected with wild type dog rabies virus (RABV, wt DRV-Mexico) correlates with the presence of high level of virus neutralizing antibodies (VNA) in the cerebral spinal fluid (CSF) and mild immune cell accumulation in the central nervous system (CNS). By contrast, dogs that succumbed to rabies showed only little or no VNA in the serum or in the CSF and severe inflammation in the CNS. Dogs vaccinated with a rabies vaccine showed no clinical signs of rabies and survived challenge with a lethal dose of wild-type DRV. VNA was detected in the serum, but not in the CSF of immunized dogs. Thus the presence of VNA is critical for inhibiting virus spread within the CNS and eventually clearing the virus from the CNS.

Conclusions/Significance

Non-lethal infection with wt RABV correlates with the presence of VNA in the CNS. Therefore production of VNA within the CNS or invasion of VNA from the periphery into the CNS via compromised blood-brain barrier is important for clearing the virus infection from CNS, thereby preventing an otherwise lethal rabies virus infection.     

Immune sera and antiglycoprotein monoclonal antibodies inhibit in vitro cell-to-cell spread of pathogenic rabies viruses.

Although the cell-to-cell spread of many viruses in vitro is inhibited by antibody, the effect of antibody on such spread of rabies viruses is uncertain. Thus, we examined the effects of anti-rabies virus immune sera and monoclonal antibodies (MAbs) on the in vitro spread of pathogenic rabies viruses in neuronal and nonneuronal cells. Both anti-rabies virus immune sera and neutralizing antiglycoprotein MAbs inhibited the cell-to-cell spread of street rabies virus, challenge virus standard, and ERA rabies viruses in cultures of neuroblastoma cells and of nonneuronal BHK-21 and chicken embryo-related cells. Furthermore, the cell-to-cell spread of virus was inhibited by greater than or equal to 75% with less than 1 IU/ml of human antirabies immunoglobulin. Nonneutralizing antinucleocapsid MAbs did not inhibit viral spread. After the immune serum was removed from the monolayers, virus spread rapidly to uninfected cells. Thus, antibody controlled the cell-to-cell spread of the virus but did not eliminate it from the cultures. Because antibody was more effective in inhibiting viral spread in fibroblast and epithelioid cells than in neuroblastoma cells infected at a high multiplicity of infection, we suggest that the inhibition of viral cell-to-cell spread by antibody in vivo would more likely occur at an initial site of exposure and before nerves are infected.     

Attenuated rabies virus activates, while pathogenic rabies virus evades, the host innate immune responses in the central nervous system

Rabies virus (RV) induces encephalomyelitis in humans and animals. However, the pathogenic mechanism of rabies is not fully understood. To investigate the host responses to RV infection, we examined and compared the pathology, particularly the inflammatory responses, and the gene expression profiles in the brains of mice infected with wild-type (wt) virus silver-haired bat RV (SHBRV) or laboratory-adapted virus B2C, using a mouse genomic array (Affymetrix). Extensive inflammatory responses were observed in animals infected with the attenuated RV, but little or no inflammatory responses were found in mice infected with wt RV. Furthermore, attenuated RV induced the expression of the genes involved in the innate immune and antiviral responses, especially those related to the alpha/beta interferon (IFN-alpha/beta) signaling pathways and inflammatory chemokines. For the IFN-alpha/beta signaling pathways, many of the interferon regulatory genes, such as the signal transduction activation transducers and interferon regulatory factors, as well as the effector genes, for example, 2'-5'-oligoadenylate synthetase and myxovirus proteins, are highly induced in mice infected with attenuated RV. However, many of these genes were not up-regulated in mice infected with wt SHBRV. The data obtained by microarray analysis were confirmed by real-time PCR. Together, these data suggest that attenuated RV activates, while pathogenic RV evades, the host innate immune and antiviral responses.     

Proteomics analysis of BHK‐21 cells infected with a fixed strain of rabies virus

Rabies is a neurotropic virus that causes a life threatening acute viral encephalitis. The complex relationship of rabies virus (RV) with the host leads to its replication and spreading toward the neural network, where viral pathogenic effects appeared as neuronal dysfunction. In order to better understand the molecular basis of this relationship, a proteomics study on baby hamster kidney cells infected with challenge virus standard strain of RV was performed. This cell line is an in vitro model for rabies infection and is commonly used for viral seed preparation. The direct effect of the virus on cellular protein machinery was investigated by 2‐DE proteome mapping of infected versus control cells followed by LC‐MS/MS identification. This analysis revealed significant changes in expression of 14 proteins, seven of these proteins were viral and the remaining were host proteins with different known functions: cytoskeletal (capping protein, vimentin), anti‐oxidative stress (superoxide dismutase), regulatory (Stathmin), and protein synthesis (P0). Despite of limited changes appeared upon rabies infection, they present a set of interesting biochemical pathways for further investigation on viral‐host interaction.     

Collaboration of Antibody and Inflammation in Clearance of Rabies Virus from the Central Nervous System

To investigate the involvement of various cellular and humoral aspects of immunity in the clearance of rabies virus from the central nervous system, (CNS), we studied the development of clinical signs and virus clearance from the CNS in knockout mice lacking either B and T cells, CD8⁺ cytotoxic T cells, B cells, alpha/beta interferon (IFN-α/β) receptors, IFN-γ receptors, or complement components C3 and C4. Following intranasal infection with the attenuated rabies virus CVS-F3, normal adult mice of different genetic backgrounds developed a transient disease characterized by loss of body weight and appetite depression which peaked at 13 days postinfection (p.i.). While these animals had completely recovered by day 21 p.i., mice lacking either B and T cells or B cells alone developed a progressive disease and succumbed to infection. Mice lacking either CD8⁺ T cells, IFN receptors, or complement components C3 and C4 showed no significant differences in the development of clinical signs by comparison with intact counterparts having the same genetic background. However, while infectious virus and viral RNA could be detected in normal control mice only until day 8 p.i., in all of the gene knockout mice studied except those lacking C3 and C4, virus infection persisted through day 21 p.i. Analysis of rabies virus-specific antibody production together with histological assessment of brain inflammation in infected animals revealed that clearance of CVS-F3 by 21 days p.i. correlated with both a strong inflammatory response in the CNS early in the infection (day 8 p.i.), and the rapid (day 10 p.i.) production of significant levels of virus-neutralizing antibody (VNA). These studies confirm that rabies VNA is an absolute requirement for clearance of an established rabies virus infection. However, for the latter to occur in a timely fashion, collaboration between VNA and inflammatory mechanisms is necessary.     

The multibasic cleavage site in H5N1 virus is critical for systemic spread along the olfactory and hematogenous routes in ferrets

The route by which highly pathogenic avian influenza (HPAI) H5N1 virus spreads systemically, including the central nervous system (CNS), is largely unknown in mammals. Especially, the olfactory route, which could be a route of entry into the CNS, has not been studied in detail. Although the multibasic cleavage site (MBCS) in the hemagglutinin (HA) of HPAI H5N1 viruses is a major determinant of systemic spread in poultry, the association between the MBCS and systemic spread in mammals is less clear. Here we determined the virus distribution of HPAI H5N1 virus in ferrets in time and space-including along the olfactory route-and the role of the MBCS in systemic replication. Intranasal inoculation with wild-type H5N1 virus revealed extensive replication in the olfactory mucosa, from which it spread to the olfactory bulb and the rest of the CNS, including the cerebrospinal fluid (CSF). Virus spread to the heart, liver, pancreas, and colon was also detected, indicating hematogenous spread. Ferrets inoculated intranasally with H5N1 virus lacking an MBCS demonstrated respiratory tract infection only. In conclusion, HPAI H5N1 virus can spread systemically via two different routes, olfactory and hematogenous, in ferrets. This systemic spread was dependent on the presence of the MBCS in HA.